SLGPT: Using Transfer Learning to Directly Generate Simulink Model Files and Find Bugs in the Simulink Toolchain

TL;DR

SLGPT employs transfer learning with GPT-2 to generate Simulink models and detect bugs, outperforming existing methods by producing more realistic models and identifying more bugs in the toolchain.

Contribution

This paper introduces SLGPT, a novel transfer learning approach that adapts GPT-2 for generating Simulink models and bug detection, leveraging large pre-trained models and open-source data.

Findings

SLGPT generates models closer to real open-source models than competitors.

SLGPT finds a broader set of bugs in the Simulink toolchain.

SLGPT outperforms DeepFuzzSL in model similarity and bug detection.

Abstract

Finding bugs in a commercial cyber-physical system (CPS) development tool such as Simulink is hard as its codebase contains millions of lines of code and complete formal language specifications are not available. While deep learning techniques promise to learn such language specifications from sample models, deep learning needs a large number of training data to work well. SLGPT addresses this problem by using transfer learning to leverage the powerful Generative Pre-trained Transformer 2 (GPT-2) model, which has been pre-trained on a large set of training data. SLGPT adapts GPT-2 to Simulink with both randomly generated models and models mined from open-source repositories. SLGPT produced Simulink models that are both more similar to open-source models than its closest competitor, DeepFuzzSL, and found a super-set of the Simulink development toolchain bugs found by DeepFuzzSL.